AudioMuse-AI-NV-plugin

AudioMuse-AI Navidrome Plugin

AudioMuse-AI-NV-Plugin the a Navidrome plugin that integrates core AudioMuse-AI features into the Navidrome frontend.

Actually this is the list of integrated functionality:

• Instant Mix - Song similarity
• Radio - Artist Similarity
• Artist Info - It return similar artist

For Mobile app that want to map this functionality they need to implement the getSimilarSongs2 / getSimilarSongs and getArtistInfo API.

Front-End tested with the plugin are:

• Navidrome integrated web frontend
• Substreamer - iOS/Android opensource mobile frontend, more information here: https://github.com/ghenry22/substreamer/issues/58
• Tempus - Android opensource mobile frontend, more information here: https://github.com/eddyizm/tempus/issues/410
• Symfonium - Androind closed source mobile frontend. You need to enable in the configuration Use similar tracks for Radio Mix. Additional functionality that use AudioMuse-AI API are actually implemented only using Jellyfin. More information here: https://support.symfonium.app/t/implement-navidrome-audiomuse-ai-plugin-to-symfonium/12238/12
• Feishin - Web opensource frontend, more information here: https://github.com/jeffvli/feishin/issues/1675
• Wavio - Android opensource mobile frontend, more information here: https://github.com/Joel-Mercier/wavio

Other frontnend not in this list could also work by using those API even.

NEWS

• From AudioMuse-AI-NV-plugin v12 InstantMix support also Song Alchemy. It send only one song but it give you the possibility to add randomness
• From AudioMuse-AI-NV-plugin v11 InstantMix support also DCLAP. To use it you must have AudioMuse-AI v3.7.0 or higher
• From AudioMuse-AI-NV-plugin v10 different AudioMuse-AI functionality can be selected for the InstantMix, chose it in the plugin configuration
• From AudioMuse-AI-NV-plugin v8 the Sonic Similarity API extension are supported on top of the previous one. The API are documented here and supported by this Navidrome PR
• InstantMix support in Navidrome start from v0.60.0: https://github.com/navidrome/navidrome/releases/tag/v0.60.0

The full list or AudioMuse-AI related repository are:

• AudioMuse-AI: the core application, it run Flask and Worker containers to actually run all the feature;
• AudioMuse-AI Helm Chart: helm chart for easy installation on Kubernetes;
• AudioMuse-AI Plugin for Jellyfin: Jellyfin Plugin;
• AudioMuse-AI Plugin for Navidrome: Navidrome Plugin;
• AudioMuse-AI MusicServer: Open Subsonic-like Music Server with integrated sonic functionality.

HOW-TO Install

IMPORTANT: Before start we suggest to have the last version of Navidrome, AudioMuse-AI-NV-plugin and AudioMuse-AI core containers. If the version are not aligned, some error could happen. Keep this in mind also for future update.

• The ENV var ND_PLUGINS_ENABLED, ND_PLUGINS_AUTORELOAD and ND_AGENTS are important, assuming that you deploy with docker compose you should use something like this:

version: '3'
services:
  navidrome:
    image: deluan/navidrome:latest
    ports:
      - '4533:4533'
    environment:
      - ND_PLUGINS_ENABLED=true
      - ND_PLUGINS_AUTORELOAD=true
      - ND_AGENTS=audiomuseai,lastfm,deezer
      - ND_DEVARTISTINFOTIMETOLIVE=1s
    volumes:
      - ./data:/data
      - /path/to/music:/music:ro

• Then you need to put audiomuseai.ndp in Navidrome data plugins folder (default: /data/plugins).
• Restart Navidrome, go to UI -> Plugins, enable AudioMuse-AI, set AudioMuse-AI API URL and other configuration parameter.
• The order of ND_AGENTS is important. Navidrome will use the first listed agent supporting sonic similarity.

[!IMPORTANT]

• The audiomuseai.npd can be found attached to the release: https://github.com/NeptuneHub/AudioMuse-AI-NV-plugin/releases.
• If you had configured authentication on AudioMuse-AI, you should also to create an 'API Token' on AudioMuse-AI core container and put in the plugin configuration.
• If AudioMuse-Ai has multiple Music Server configured, you need to configure the 'Server Name' in the plugin to match the one in AudioMuse-AI
• The Instant Mix Functionality select choose which AudioMuse-AI engine answer Instant Mix and Sonic Similarity:
  • Similar Song (default) - GET /api/similar_tracks, the historical behaviour;
  • Lyrics by Song - POST /api/sem_grove/search, it need the SemGrove index (lyrics + audio analysis) built in AudioMuse-AI;
  • Hyperbolic Similarity - POST /api/hyperbolic/similar in its standard mode, it need the hyperbolic projection built in AudioMuse-AI;
  • DCLAP - POST /api/clap/search_by_song, it need CLAP enabled and the DCLAP index built in AudioMuse-AI;
  • Song Alchemy - POST /api/alchemy with the song as the only ADD item and the configured 'Temperature' (default 1), it's the MusiCNN similarity with randomness.
  'Eliminate Duplicates' apply to Similar Song and DCLAP, 'Use Radius Similarity' apply only to Similar Song, the other engine use the AudioMuse-AI default.

See the official Navidrome Documentation for more information on how the plugin works.

HOW-TO test

In the AudioMuse-AI integrated UI, use the Similar Song functionality on a track.

In the Navidrome Web UI, use InstantMix on the same track.

Both actions should generate similar requests in the AudioMuse-AI Flask logs. For the Navidrome request, you should also see plugin=audiomuseai in multiple line of the Navidrome logs with no related errors.

If both tests pass, supported third-party clients should work correctly. Any issue on third party client MUST be addressed on their side.

[!NOTE]

• Different frontends may return a different number of similar songs, depending on their configuration. This plugin does not enforce a default limit.
• A 401 Unauthorized error in the Navidrome logs typically indicates that the apiToken is missing or incorrectly configured.
• Navidrome must be able to reach the AudioMuse-AI core container. Ensure the plugin is configured with the correct host/IP address. If requests do not appear in the AudioMuse-AI Flask logs, the issue is most likely network connectivity or container routing.

HOW-TO build

• Requirements (Ubuntu / macOS): Go, TinyGo.
• Build:

make build    # -> audiomuseai.wasm
make package  # -> audiomuseai.ndp

Full stop.

Code Mirror

AudioMuse-AI-NV-plugin repository code is mirrored here:

• https://codeberg.org/NeptuneHub/AudioMuse-AI-NV-plugin

DO NOT USE MIRROR TO RAISE ISSUE, PR OTHER ACTION DIFFERENT FROM GET THE CODE